Code Review Checklist

Systematic review criteria for evaluating code and plans. Use this skill when:

• Critic reviews plans before implementation
• Security agent conducts code audits
• Architect reviews architectural compliance
• UAT validates implementation quality

Review Context

This skill supports two review phases:

Phase	Agent	Focus	Documents
Pre-Implementation	Critic	Plan quality, clarity, completeness	planning/*.md
Post-Implementation	Security, Architect	Code quality, security, architecture	Source code

---

Pre-Implementation Review (Critic)

Value Statement Assessment (MUST START HERE)

Check	Question	Finding Severity
Presence	Does plan have clear value statement in user story format?	CRITICAL if missing
Clarity	Is “So that” outcome measurable or verifiable?	HIGH if vague
Alignment	Does value support Master Product Objective?	CRITICAL if drift
Directness	Is value delivered directly, not deferred?	HIGH if deferred

Plan Completeness

Check	Question	Finding Severity
Scope	Are boundaries clearly defined?	MEDIUM
Deliverables	Are all deliverables listed with acceptance criteria?	HIGH
Dependencies	Are dependencies identified and sequenced?	MEDIUM
Risks	Are risks documented with mitigations?	LOW
Version	Is semver bump specified with rationale?	MEDIUM

Constraint Compliance

Check	Question	Finding Severity
No Code	Does plan avoid prescriptive code?	LOW
No How	Does plan focus on WHAT/WHY, not HOW?	LOW
Architecture	Does plan respect architectural constraints?	HIGH

---

Post-Implementation Review (Security/Architect)

Security Checklist

Category	Check	Severity
Input Validation	All user input validated server-side?	CRITICAL
Authentication	Auth checks on all protected routes?	CRITICAL
Authorization	RBAC/ownership verified before access?	CRITICAL
Secrets	No hardcoded credentials or keys?	CRITICAL
SQL/Injection	Parameterized queries used?	CRITICAL
XSS	Output encoding applied?	HIGH
CSRF	Tokens on state-changing requests?	HIGH
Logging	Security events logged without sensitive data?	MEDIUM
Dependencies	No known CVEs in dependencies?	Varies

Performance Checklist

Category	Check	Severity
N+1 Queries	Batch fetches instead of loops?	HIGH
Pagination	Large datasets paginated?	HIGH
Caching	Appropriate caching strategy?	MEDIUM
Async	Long operations non-blocking?	MEDIUM
Resource Limits	Bounded allocations?	HIGH

Maintainability Checklist

Category	Check	Severity
Naming	Clear, descriptive names?	LOW
Complexity	Cyclomatic complexity < 10?	MEDIUM
Coupling	Low coupling between modules?	MEDIUM
Documentation	Public APIs documented?	LOW
Error Handling	Errors handled, not swallowed?	HIGH
Tests	Adequate coverage for changes?	HIGH

Architectural Compliance

Category	Check	Severity
Boundaries	Module boundaries respected?	HIGH
Patterns	Established patterns followed?	MEDIUM
Dependencies	Dependency direction correct?	HIGH
Single Responsibility	Classes/modules focused?	MEDIUM

---

Severity Definitions

Severity	Response	Examples
CRITICAL	Block until fixed	Auth bypass, SQL injection, no value statement
HIGH	Fix before merge	Missing validation, N+1 queries, unclear scope
MEDIUM	Fix in current cycle	Code smells, missing docs, minor coupling
LOW	Track for later	Style issues, optimization opportunities

---

Finding Format

Document findings consistently:

### [ID]: [Brief Title]
- **Severity**: CRITICAL / HIGH / MEDIUM / LOW
- **Status**: OPEN / ADDRESSED / RESOLVED / DEFERRED
- **Location**: [file:line or plan section]
- **Description**: [What is the issue?]
- **Impact**: [Why does this matter?]
- **Recommendation**: [How to fix?]

---

Agent-Specific Guidance

For Critic Agent

• Focus on plan quality, not implementation details
• Value statement assessment is mandatory first step
• Reference Planner constraints when reviewing
• Create critique in agent-output/critiques/

For Security Agent

• Focus on OWASP Top 10 and injection patterns
• Reference security-patterns skill for detection
• Create audit in agent-output/security/
• Use CVSS-aligned severity

For Architect Agent

• Focus on system-level design compliance
• Reference architecture-patterns skill
• Update system-architecture.md when issues found
• Include ADR updates if decisions affected